Définir les zones périurbaines: liens avec les milieux ruraux et urbains et cadres institutionnels
Les auteurs du présent article proposent une définition et l'ébauche d'une typologie de l'espace périurbain en fonction de ses articulations avec les espaces ruraux et urbains. Ils sont ainsi amenés à distinguer cinq types d'espace périurbain (PU): le village PU, l'espace PU diffus, la chaîne PU, l'espace PU in situ et l'espace PU phagocyté . Ces distinctions tiennent compte des processus sociodémographiques, sous-jacents, notamment des migrations.
Cette nouvelle typologie permet de mieux situer le cadre institutionnel et les réseaux opérant dans chaque type d'espace périurbain. Ainsi, les agents du développement peuvent utiliser cet instrument pour identifier les institutions clés dans leur zone d'activité. À titre d'exemple, les auteurs appliquent ce cadre d'analyse aux modalités de la propriété foncière et de l'héritage. Ils proposent comme second exemple le vieillissement de la population, sous forme d'un tableau présenté sans commentaires. Enfin, ils évoquent certaines interrogations et difficultés qui ne sont pas encore résolues à ce jour.
Una definición de «periurbano»: relaciones entre el sector rural y el urbano y vinculaciones institucionales
En este artículo los autores proponen una definición de «periurbano» y elaboran una útil tipología para este concepto que tiene en cuenta sus relaciones con los entornos rural y urbano. Se identifican concretamente cinco tipos de sectores periurbanos: de aldeas, dispersos, vinculados, localizados y absorbidos. Esta tipología deriva de los procesos sociodemográficos subyacentes, en particular la migración.
La nueva definición ayuda a identificar el marco institucional y las redes pertinentes en las distintas áreas periurbanas, de manera que los trabajadores de desarrollo pueden utilizarla para identificar las instituciones clave en sus áreas de interés. A título de ejemplo los autores aplican este marco de referencia a la esfera de las normas de tenencia/herencia de la tierra. Aunque incluyen un segundo ejemplo sobre el envejecimiento en el cuadro recapitulativo, este tema no se analiza en el artículo. Los autores señalan, por último, algunos de los problemas y limitaciones que aún quedan por resolver.
D.L. Iaquinta and A.W. Drescher
David L. Iaquinta can be contacted at Nebraska Wesleyan University, Sociology/Anthropology/Social Work, 5000 Saint Paul Avenue, Lincoln, Nebraska 68504-2796, United States; e-mail: [email protected] Axel W. Drescher can be contacted at University of Freiburg, Section on Applied Physiogeography of the Tropics and Subtropics (APT), Hebelstrasse 27; 79104 Freiburg, Germany; e-mail: [email protected]
The authors of this article propose a definition of peri-urban and elaborate a conceptual peri-urban typology, including its relationships to rural and urban forms. Specifically, they identify five peri-urban (PU) types: village PU, diffuse PU, chain PU, in-place PU and absorbed PU. The typology derives from underlying sociodemographic processes, especially migration.
The new definition helps to identify the institutional framework and relevant networks in the different peri-urban areas. Thus, development workers can use it as a tool to identify the key institutions in their area(s) of interest. As an example, the authors apply the framework to the area of land tenure/inheritance rules. They include a second example, based on the ageing of a population, but do not discuss it at length. The article finishes with identification of some of the, as yet, unresolved issues and constraints.1
This article is a response to the growing interest in food production in urban and peri-urban environments (urban and peri-urban agriculture [UPA]).2Increasingly, policy-makers and researchers are acknowledging the potential role that urban and peri-urban environments play in alleviating food insecurity and enhancing the nutritional status of urban poor and marginalized people (FAO, 1999a). As UPA itself has become more prevalent, so too has awareness of its potential and limitations.
There is an increasing perception that rural, peri-urban and urban environments operate as a system rather than independently. Many development specialists conclude that rural development and urban planning are necessarily linked activities. Activities or interventions in one arena have consequences, which are often negative, in the other. At the same time, creative policies can turn liabilities into resources and bridge the rural-urban divide.
Problems that are thought to be biological or technical often have their roots in sociocultural arrangements and processes. Similarly, problems that are manifested in one geographic environment often originate in a different environment. Both urban and rural development policies have impacts on peri-urban environments. In some cases, these impacts can lead to the creation of new peri-urban sites. In fact, a key feature of peri-urban environments is their dynamic nature, wherein social forms and arrangements are created, modified and discarded. They are areas of social compression or intensification where the density of social forms, types and meanings increases, fomenting both conflict and social evolution.
Unfortunately, research and policy discussion regarding UPA has been hampered by:
This paper is concerned with the definition of peri-urban and the elaboration of a conceptual typology. The work rests on six premises:
The term peri-urban is used frequently in the literature and in policy discussions, yet definitions are largely situational and case-specific. They provide little basis for a unified understanding of what constitutes peri-urban. This point was made by the Organisation for Economic Co-operation and Development (OECD) in its report on peri-urban agriculture (OECD, 1979: 10) which states:
The term "peri-urban area", cannot be easily defined or delimited through unambiguous criteria. It is a name given to the grey area which is neither entirely urban nor purely rural in the traditional sense; it is at most the partly urbanized rural area. Whatever definition may be given to it, it cannot eliminate some degree of arbitrariness."
The goal of this article is to bring some theoretical clarity and practical utility to this problem. First, the essential points of agreement and the range of differing viewpoints are abstracted from the various definitions of peri-urban found in the literature. Next, a typology of peri-urban is elaborated, including its relationships to rural and urban forms. In conclusion, the elements of the typology are linked to their implied institutional contexts and important constraints and opportunities for exploiting the typology in the field are suggested.
Many rural development efforts have failed because the wrong institutions have been addressed. Addressing, stimulating and promoting the right institutions has become a key issue in participatory development work. The new definition described in this article helps to identify the institutional framework and relevant networks for different peri-urban areas. Development workers can therefore use it as a tool for identifying the key institutions in their area(s) of interest.
Although it has been in use for over half a century,3 theorists and practitioners have not been especially clear or consistent in their use of the term peri-urban. Today, researchers from many disciplinary and paradigmatic perspectives use it to describe seemingly contradictory processes and environments. Definitions are based on a variety of operational variables, and a variable that is seen as definitional by some authors is seen as being an outcome of peri-urban processes by others. Consequently, the concept of peri-urban has become trivialized and tautological, severely compromising its analytical and practical utility.
Examination of the literature shows a number of distinct patterns in the way in which researchers have addressed the issue of peri-urban. Each of these patterns, and the conclusions that can be drawn from it, is summarized in this section of the article and discussed in more detail in FAO, 1999a.
Many authors make no attempt to define what peri-urban means or to cite a source for the term, yet they use peri-urban as a substantive category or phenomenon in their work (e.g. Clough, 1996). The following conclusions can be drawn from this "implicit definition":
Other authors undertake a critical review of the theoretical concept, but subsequently employ an exclusively urban fringe operational definition,4 leading to the following conclusions:
A third category of research defines peri-urban only in relation to specific uses, markets or factors (e.g. Holland et. al., 1996). From this perspective the following can be deduced:
A fourth and final pattern in the literature comprises those bodies of work that identify the essential components of a broad-based operational definition. As OECD (1979: 9) states:
The impacts of economic growth and physical expansion of the urban area are not confined within urban boundaries; they reach into much wider areas surrounding urban centres, creating so-called "rurban areas", "urban fringe areas", or "peri-urban areas". While the peri-urban area retains the characteristics of the rural area, these are subject to major modifications: changes take place with respect to physical configuration, economic activities, social relationships and so forth.
These same three dimensions - physical configuration, economic activities and social relationships - have long existed in the literature that defines what it is to be urban or rural.
The importance of the concept of peri-urban emerged as a result of limitations in the dichotomy between rural and urban.5 Much research has identified the inadequacy of this simplistic dichotomy,6 some authors even suggesting that it no longer has relevance to social analysis.7 Others have argued more specifically that only the dichotomous construct has outlived its usefulness, not the underlying distinction between degrees of ruralness and degrees of urbanness (Rambaud, 1973).
Researchers at the Rural Policy Research Institute (RUPRI, 1998) identify the following four key limitations of the definitions of urban and rural areas (see also United States Census Bureau, 1999):
Such a position implies that rural-urban should be viewed as a continuum - albeit one that is not necessarily smooth or one-dimensional - within which individuals, households, communities and institutions are distributed. Accepting the continuum model, rather than a model that employs a "typological set", is less important than recognizing that the spectrum of change is discontinuous and multidimensional, and that it arises from underlying social processes.
The search for a comprehensive definition of peri-urban can start from the concept of urban.8 Established theoretical definitions of urbanization and urbanism identify the following three components:
In sociology, the first two components are usually taken as the basis for defining urbanization, while the third is the core definition of urbanism - the social-psychological reflection or response to urbanization (Fischer, 1977 and 1984). The social-psychological component essentially refers to those values, attitudes, tastes and behaviours that are seen to be characteristic of urban as opposed to rural dwellers. Earlier notions of this component tended to be associated with Westernization, reflecting the ethnocentric thinking of the time (Holleman, 1964: 324). Recent writers take a more neutral view of the process. For example, adapting the ideas of Friedman and Wolff (1982), Williams, Brunn and Darden (1983: 5) define urbanization as a:
process involving two phases or aspects: i) the movement of people from rural to urban places where they engage primarily in non-rural functions or occupations; and ii) the change in life style from rural to urban with its associated values, attitudes and behaviours. The important variables in the former are population density and economic functions; the important variables of the latter depend on social, psychological and behavioural factors. The two aspects are mutually supportive.
Given that the three components form the basis of the definition of urban and given that peri-urban includes some level of urbanness, it follows that some variation of the three components should underlie the definition of peri-urban. Holleman underlines this when he links peri-urban directly to the concept of a "change in mental orientation."9
It is important that "proximity to the city" seems not to be essential to the definition of peri-urban. The fact that many peri-urban areas are close to the city is substantively important and instrumental to a comprehensive understanding of peri-urban, but it is incidental to an elemental understanding of peri-urban. "Proximity to city" represents a further specification, which allows distinctions to be drawn among the types of peri-urban but does not define peri-urban per se. In addition, concentration on geographic location as a basis for defining peri-urban also undermines a clear understanding of the rural-urban spectrum as dynamic, interactive and transformative.
The social-psychological component is the one most often omitted from peri-urban definitions. Yet, as some scholars and policy analysts have argued, to ignore this component is to overlook the full meaning of peri-urban, underestimate the prevalence of social change and misclassify the experiences of numerous people and communities.10 Typically, the researchers who have supported inclusion of this component have also been the most critical of definitions that rely solely on proximity to the city.11
Urbanization is a process of concentration and intensification of human life and activity. It is an uneven process that takes place in a physical environment. Urbanization is one possible outcome of the three fundamental population processes: fertility, mortality and migration. These processes are, in turn, the results of individual decisions undertaken within a sociocultural, economic, political and environmental context. One consequence of urbanization is the uneven incorporation of a variety of institutional forms into larger cultural environments which are identified as urban, peri-urban and rural. In particular, five classes of institutional arrangements can be identified as arising within the complex continuum from rural to urban and falling within the range of phenomena that scholars and practitioners have identified as peri-urban. Each of these institutional classes is connected to a specific peri-urban type and hypothesized to arise from a specific demographic process (in parentheses in the following) underlying urbanization:
The importance of these classes of institutional arrangements is that they can help to identify useful meso-policy interventions. This is important in urban and peri-urban environments where there is an intensification of conflict and the necessity to negotiate and resolve competing claims (e.g. land for residential or agricultural use, or diverse customary institutional forms and values) and implement development plans. Such conflicts occur at all levels, including the family, the neighbourhood, the community, the local, the regional and the national.
The nature and complexity of the society that combines elements of both rural and urban is well established in the literature. It has been described under a number of guises and given various definitions by different researchers. Terms such as peri-urban, ex-urban, urban tract, rurban, urban fringe, semi-urban and even suburban have all been used variously to describe environments that are in some way neither rural nor urban and yet embody elements of both. See Anderson (1914: 85) for an early example of this thinking.
Village PU (or peri-rural): "rural places with urban consciousness". This category refers to areas that are geographically not close to an urban area, yet are experiencing substantial urbanism (i.e. the social-psychological dimension of urbanization). While such influences can accrue solely through mass media and the diffusion of consumerist ideologies, in developing countries they are more likely to occur as a result of such processes as:
This is the category that is most often omitted from considerations of peri-urban environments. In essence, its designation as peri-urban rests on its social-psychological transformation rather than its geography or size; the transformation itself resulting from the demographic process of migration. However, rather than focusing on the geographic movement of out-migrants, it is important to emphasize the continuing linkages by which they effect the infusion of the urban into the village culture.12 It is significant that these environments are likely to be very stable yet capable of absorbing and accommodating urban values. The mechanism of such accommodation depends on the stability of the community and the structured network of out-migrant participation.13
Diffuse PU: in-migration from several places. A separate category of peri-urban is comprised of areas proximate to the city that have been settled through in-migration. In this instance, the in-migrants derive from a variety of geographic source points rather than a single one. In-migration to these environments often also includes migrants from urban areas. These areas are characterized by greater ethnic heterogeneity and greater variety of beliefs about customary institutions and arrangements than chain PU environments. The institutional patterns here reflect far greater inclusion of urban forms than is the case for either chain PU or in-place PU.
Diffuse PU environments have greater potential than chain PU environments for both the outbreak of conflict and the negotiation of new institutions that are more urban-oriented. Such areas of settlement may arise from a "staged" occupation, whereby unoccupied land is settled by landless people during a coordinated take-over at a time specific (cf. de Soto, 1989). These areas may also arise from spontaneous processes of migration over time, whereby people from diverse origins - mostly the poor and landless - settle together.14 It is important to remember that the heterogeneity of the cultures of origin requires that any collective organization must be negotiated across, rather than along, customary lines.15 Simple adherence to tradition is insufficient to settle conflicts that derive directly from differences between traditions. There is therefore a need for increased use of modern (i.e. urban) or transcultural modes and methods of dispute resolution and community building that transcend particular traditions. The likelihood of such cross-cultural negotiation is increased by the in-migration of urban residents, whether they are long-term urban-dwellers or more recent in-migrants from rural areas temporarily making use of urban ghettos for migration.16
Chain PU: in-migration from a single place. Some areas proximate to the city undergo settlement through a process of chain migration, i.e. the geographic translocation of a village population to a specific locality in the urban periphery. These migrants tend to be the most opportunistic (in terms of taking risks) members of their original village populations, hence the most accepting of change. Chain PU areas have a high degree of ethnic homogeneity and population numbers that are sufficient for a critical mass. Consequently, traditional/customary beliefs and institutions tend to be transferred and reconstructed in the new environment, integrated with elements from the surrounding urban institutions.17 Such integration of urban institutions happens to a greater extent in chain PU than in-place PU areas.
This type of community formation is similar to that described by Gans (1962) as leading to the creation of "urban villages". Indeed, chain migration is the master trend underlying much international migration. Early migrants, or "pioneers", serve as the forerunners for later migrations of settlers from the homeland. By providing temporary housing and information about the ways of the new culture, the pioneers reinforce their status as landsmann. This process also reinforces the tendencies both to form enclaves and to reproduce adapted traditional institutions, sometimes along the lines of kinship and sometimes along the lines of landsmannschaften or coethnicity. This type of peri-urban community is highly stable. Areas identified as "squatter settlements" around the cities of developing countries are mostly this type or diffuse peri-urban.
In-place PU. These areas are proximate to the urban area and result from in-place (in situ) urbanization. That is, they are in the process of being absorbed into the urban environment, whether by annexation (expansion of the city fringe) or simple reclassification (reflecting de facto urban expansion). In some instances, in-place PU areas become more urban through natural expansion and/or rural in-migration. More commonly, they are formed from peri-urban villages combined with in-migration from the nearby urban area.18 Whichever is the case, because they are being absorbed "whole", such places tend to perpetuate and reinforce the existing power structure and bases of inequality. To the degree that sufficiently large numbers of migrants into the area arrive from the city, conflicts between established residents and newcomers are likely to emerge.19 Excluding newcomers, the residents of these areas tend to reflect the extremes of the local power spectrum, i.e. they are those who are:
Because of their lack of geographic displacement and the potential for increased polarization between established residents and newcomers, these environments are likely to have the most intact and conservatively held customary and traditional institutions.
Absorbed PU. The final category of peri-urban refers to areas that have been proximate to or within the urban context for a considerable period. The defining characteristic of these locations is the maintenance of customary or traditional institutional arrangements that are derived from the culture of original settlers/residents who have long since ceased to be the numeric majority in the area. Such areas derive from either in-place PU or chain PU areas. Over time, either of these peri-urban types can undergo the compositional processes of succession and displacement, while on the macrolevel being increasingly absorbed into the urban environment - administratively, politically and social-psychologically.
In short, the original settler culture group is replaced through either residential succession or diffusion caused by differential migration along ethnic/cultural lines. However, some important customary arrangements (i.e. institutions) of the original group remain in place, now supported by newcomers, through a combination of ritualism, power/dominance relations and reification by arrangements in the formal/modern sector. Such mechanisms have a strong conservative effect through adherence to tradition for tradition's sake rather than adherence to traditional principles because they are functional for the community.20
The search for a definition of peri-urban raises the larger question of the relationship between rural and urban environments. One observation that is well established in the literature is that rural out-migrants generally do not go directly to large cities. Rather, a series of moves are involved, wherein rural migrants move first to villages or small towns and successively to more urban environments. A second observation in the literature is that migration does not sever all, or even most, of the linkages between the migrant and her/his community of origin and family.21
Together these two points emphasize the importance of conceptualizing the peri-urban environment as a dynamic, transformative and reciprocal arena linked at the macrolevel not only by economic activities and geography but also, and significantly, by the social fabric of individual and family networks.22 Thus, the peri-urban environment is dynamic exactly because of the flow of migrants and the density and heterogeneity of activities present. It is transformative because it changes the migrants and the migrants change it. It is reciprocal, not only because individual migrants and the social environment influence each other, but also because the individual links between donor areas and receiving areas continue to transmit change in both directions at the aggregate and institutional levels.23
These comments suggest a further elaboration of the peri-urban types discussed in the previous section. Two kinds of links can be identified at the macrolevel: links that persist across space in the face of geographic displacement; and links that persist across time. The Figure captures the dynamics of both types of links. The links can be conceptualized in terms of the peri-urban types themselves or in terms of the institutional contexts that they imply. Both conceptual schemes are included in the Figure, and the institutional contexts are elaborated in the next section of this article.
In the Figure, horizontal arrows represent links across space. These linkages are the direct result of migration, either chain or diffuse, that operates in a way that creates individualized exchange networks across space. The accumulation of such individual social capital links geographically distinct areas into a larger exchange network. The vertical arrows represent links across time. These linkages result from the passage of time in a given area, allowing for the accumulation of demographic, social and institutional change.24
The upper part of the Figure is shaded to isolate the portion of the model that is primarily concerned with the transformative effects of migration from the portion primarily concerned with changes over time.25 Thus, the five peri-urban types are embedded within the broader rural-urban dynamic. From this vantage point, two interacting subsystems are identified. Note that the Figure does not include arrows to identify all the possible migration flows or changes over time; rather it depicts the principal flows and mechanisms of change. However, arrows that indicate the migration flow directly from rural to urban, and its counterpart in the opposite direction, have been included. Note also that arrows showing the counter flows from urban to diffuse PU and village PU have been included.
The typology of peri-urban must now be translated into a tool that has social and analytical relevance. An important first step is to identify the institutional contexts implied by each of the elements in the typology. Such contexts appear for each peri-urban type in parentheses in the Figure. The term "institutional" reflects the broad range of cultural meanings and social organizations that encompass customary and informal relations in particular. The term "contexts" refers to the essential institutional features and structural constraints implicit in each element. Table 1 summarizes the institutional contexts and characteristics associated with each element in the peri-urban typology.
A general assessment of the relationships between each institutional context and existing stratification systems is also included. This is important since, as Sonje and Stulhofer (1995) point out, horizontal institutions encourage cooperation among social actors, while vertical ones erode cooperation. Thus, institutions embodying stratification will tend to erode cooperation. However, many institutions include both horizontal and vertical elements. Thus, the distinction between horizontal and vertical institutions can be extended to include horizontal and vertical elements within an institution. In this way, institutions are not seen as being entirely beneficial or entirely detrimental, but rather beneficial to some subgroups (who will cooperate with and support the maintenance of the institution) and detrimental to others (whose cooperation with and overall solidarity of support for a given institution will be undermined).
The (network-)induced institutional context. The network-induced institutional context is associated with village PU environments. Environments of this type are tradition-oriented and, in most respects, resemble rural villages. Population size and density are relatively low and many residents are involved in agricultural production. The key differentiating factor is the social-psychological orientation of the population. As a result of the out-migration of some residents, urban attitudes and values are introduced to the community. This process of diffusion or induction is driven by the circulation and sojourning of the out-migrants and, in general, by their maintenance of individual exchange networks with their villages of origin.
Despite the introduction of urban attitudes and tastes through the out-migrants, the institutions of the village remain traditional in orientation, and stable. New ideas, induced by out-migrant influence, are absorbed slowly into the traditional context, often through a process of redefinition. Redefinition of the situation allows for the perpetuation of an "ideal culture" in the course of adaptation to the needs of the situation. Thus, for example, village tradition may call for land to be owned, controlled and worked by men. However, a shortage of young males, caused by migration, may lead to a redefinition whereby land is still nominally owned by men but is controlled and worked by women. Such a situation may involve the shift from real decision-making by men to mere symbolic approval by men of women's decision-making.26
Even though changes are made, the traditional institutional structures remain largely intact. Because the urban ideas are brought in from outside the village, and because the village is not geographically close to the urban area, the demand for change is relatively low.
The long-term stability of the traditional system gives it high resistance to change and, thus, change is incorporated slowly. Sonje and Stulhofer (1995) attribute the stability to deeply internalized and shared informal norms that they call "sociocultural capital". They argue that institutions built on deeply internalized, tested norms will be stable, change in small evolutionary ways and resist violent, exogenous change.
Change increases the opportunities for egalitarianism and for the erosion of the gender and age stratification systems, albeit incrementally.
Amalgamated institutional context. Diffuse PU environments are formed by the influx of migrants from a variety of geographic and cultural sources. These environments lie near urban areas and also serve as migration end-points for urban out-migrants. New migrants to these areas are generally concerned with survival needs. Their compositional heterogeneity necessitates the formation of a collective identity if they are to obtain needed services from formal urban institutions. Their heterogeneity also requires them to negotiate solutions for survival and collective identity because they cannot rely on simple tradition. In fact, internally and externally conflicting cultural traditions may be a chief obstacle to functioning with the nearby urban environment.
These environments have a great need for adaptability to change owing to their proximity to the city. The influx of new migrants, the demands of coping with the nearby urban sector and the need to overcome cultural barriers require that resistance to change is low. The very selectivity of migration, whereby the innovators are most likely to have migrated, supports this low resistance to change. These environments have the greatest capacity to spawn democratic or consensus-based change and institutions. Thus, they are the environments with the greatest opportunity for egalitarianism and erosion of traditional stratification systems.
A major characteristic of this peri-urban type is the relative lack of formal institutions. Solutions generated in such an environment have to meet the needs of the modern sector and often incorporate waged labour as a significant economic component. However, the solutions are often novel. Such emergent institutional forms are an amalgam of various customary traditions and forms from the modern sector.
The chief requirement for these new systems is that they achieve some negotiated legitimacy from the participants/residents. Negotiating this legitimacy is, itself, a required first step in the process of creating new institutions in the area.
Reconstituted institutional context. This institutional context exists when an area proximate to the city becomes an end-point for the process of chain migration. Chain migration refers to the mechanism whereby early "pioneer" migrants serve as focal points for later settler migrants from the home village or area. In these environments, the dense concentration of migrants with similar cultural origins leads to the re-creation of the institutional forms that existed in the home village.
This re-creation is never exact; so the term "reconstitution" is more accurate. Reconstruction of the collective cultural identity is a defensive mechanism on the part of the new migrants who attempt to re-establish the familiar amid the alien. The situation is exacerbated by the challenge of dealing with urban formal institutions.
As with amalgamated institutional contexts, the need for change is great owing to proximity with the urban environment. However, there is also greater resistance to change in this case because of the defensive nature of the reconstituted institutions, which are organized along traditional or customary lines.
Resistance to change in the reconstituted institutional context can be classified as medium to reflect the trade-off between the conservative force of the reconstituted institutional form and the liberalizing force of migration, which drives the innovating forces in the population. Thus, change will reflect the old but include some urban/modern components, particularly those that make efficient use of the formal sector or allow for effective linkages to the modern sector. In terms of stratification systems, this institutional context reinforces those types that existed in the traditional system. However, the exact form of the stratification system may be changed.
Because of the way in which these environments are formed, individuals living in them tend to remain linked to their places of origin. Circulation and remittance flows are likely to remain significant. Thus, these environments will have a continuing impact on the more rural components of the exchange network, fostering further migration, introducing urban attitudes and values, initiating commercially valuable economic exchanges between locally produced goods and urban markets and products and, possibly, channelling capital to home areas for economic development projects.
Traditional institutional context. The processes of growth and annexation, combined with in-migration, create in-place PU environments. Unlike chain PU environments, which have benefited from risk-taking immigrants, these areas are populated by the converse of migration selectivity, namely those least likely to have migrated out of the traditional environment. Such environments generate traditional institutional contexts. While proximate to the city, they have long-term stable institutions that respond to the in-migration of other people, particularly urbanites, with defensive insulation.27
In this traditional context the need for change is medium because of the relatively stable institutional environment. Yet, this same environment creates a high resistance to change. Thus, institutional adaptation is slow and there is great potential for conflict from the increasing polarization between established residents and newcomers. Because the existing stakeholders attempt to protect their traditions through defensive insulation, the kinds of adaptations that emerge are likely to be inefficient in terms of facilitating access to the modern urban sector. The process leads to heightened conflict over control of the institutional system, and is met with increasing oppression along traditional stratification lines.
Remittance flows and circulation are of reduced importance in this context, since the environment itself is primarily intact. Thus, the impact of this environment on more rural areas is more limited than that of other peri-urban types. There are more formal institutions in this environment, but not necessarily of an urban type.
Residual institutional context (traditionalism). Residual institutional contexts are created when the original culture group has been replaced though a process of residential succession and displacement but a set of arrangements whose roots lie in the culture of the original residents has remained in place. Under such conditions, the institutional context is upheld through ritualism or traditionalism (i.e. rigid adherence to custom simply for tradition's sake, even when the basis for the tradition no longer holds) or because members of the original culture group still control the local power structure (precluding access for newcomers and rewarding compliance). These environments are actually parts of the city (i.e. they have been spatially absorbed by the city), but they are still classified as peri-urban to emphasize that the roots of their institutional arrangements lie very much in the peri-urban, rather than the urban, environment. Just as village PU environments appear rural, so do absorbed PU environments appear urban. In addition, it is primarily the social-psychological dimension that differentiates both village PU from rural environments and absorbed PU from urban environments.
Because absorbed PU environments lie within the urban environment, politically a great force for change is exerted by urban formal institutions. The residual institutional structures adapt badly to change and, ultimately, will lose their legitimacy among the residents whose needs they fail to meet. Nonetheless, there is high resistance to change, and very slow adaptation, until the community reaches a crisis of legitimization. At this point, conflict will be significant and, probably, revolutionary (whether generated internally or introduced coercively from outside by the government).
Much of the change that takes place under residual institutions will be provoked by simple compliance, in which the demands of the formal sector will be externally accepted but internally rejected. Such compliance leads to a different sort of legitimization crisis in which the formal sector's authority is undermined within the smaller absorbed PU environment. Thus, attempts to eliminate or alter the stratification system that are implicit in the residual institutions will be met with resistance. The result is increased support for the maintenance of the traditional stratification system and heightened rejection of the modern sector. A major characteristic of this peri-urban environment is the presence of both traditional and formal institutions.
Having identified various peri-urban types and the resulting institutional contexts, their relevance to important areas of social policy can now be discussed. In particular, consideration will be given to the implications of the typology in the areas of land tenure (more specifically inheritance rules) and population dynamics (more specifically population ageing).
The upper section of Table 2 summarizes the relationships between inheritance rules and the various peri-urban environments. While there are many ways of looking at the range of resources available for inheritance, here only the simple division into land and non-land wealth transmission is considered. On this basis, village PU inheritance rules are primarily oriented toward issues of land, while the rules in diffuse PU and absorbed PU environments are primarily oriented toward issues of non-land wealth transmission.
Chain PU and in-place PU environments are much more likely to involve a blend of the two types of resources, but for different reasons. For the chain PU it is because the networks that facilitate migration to the area also facilitate the accumulation of land resources, while for the in-place PU it is because the survival of customary institutions depends on the ability of elders to control the resources in the community - and first among these is land. However, for both chain PU and in-place PU areas, proximity to the urban waged labour market means that members of the community are increasingly likely to accumulate non-land wealth resources that may be transmitted to the next generation.28
The need for clear and specific rules for inheritance is particularly high in environments where land is the fundamental basis of wealth, as it is in village PU and in-place PU environments. Absorbed PU environments also need clear rules but, rather than deriving from the centrality of land, this need derives from these environments' proximity to the urban context and the prevalence of urban formal institutions. On the other hand, in diffuse PU and chain PU environments, the high concentration of poverty and general lack of formalized access to land or of formal institutions means that inheritance rules become far less necessary.
Inheritance rules are most clear and consistent in the areas where they are needed most and where the cultural context is relatively homogeneous and connected to the past, i.e. in village PU, in-place PU and absorbed PU environments. In the chain PU such rules are less clear because cultural norms are less firmly established in the amalgamated context of these environments. Nevertheless, they will be clearer in chain PU than diffuse PU environments, because the diversity of residents and cultural forms in the latter are more likely to create a highly idiosyncratic pattern of inheritance rules.
Legitimacy refers to the breadth of acceptance of a phenomenon within the population. Rules and phenomena rooted in broadly shared internalized norms will have a high degree of legitimacy.29 For inheritance rules, legitimacy will be broad and derived from custom in village PU and in-place PU environments. In the chain PU case, legitimacy is less custom-derived, owing to the greater cultural heterogeneity that is present, and it is even less custom-derived in the diffuse PU case because of the even greater heterogeneity that exists in these environments.
Structured inequality is greatest in those contexts that are most traditional in orientation, particularly in terms of age and gender issues. Thus, structured inequality is high for village PU, in-place PU and absorbed PU APU types and low for the diffuse PU. Structured inequality is moderate in chain PU environments because of the blend of traditional and modern culture.
Conflict is present in all environments, but varies in amount and source. In village PU, diffuse PU and chain PU environments, conflict is more likely to emerge along structured inequality lines, whereas in in-place PU and absorbed PU environments it is more likely to arise from the interface between customary and formal institutions. This latter type of conflict is more normative and broad in its impact, and thus more intense.
The example related to population ageing in the lower part of Table 2 is for illustrative purposes only; similar analyses can be carried out for any policy area that is distributed across the R-PU-U spectrum.
Many issues remain concerning the indicators and thresholds that differentiate rural, urban and the various peri-urban classifications. For example:
The answers to these questions vary from region to region. Yet, such variation does not limit use of the typology to one region where consistent definitions apply.
Other remaining questions are concerned with how to code the various categories. For example:
Questions also remain regarding the definition and applicability to policy-making of the institutional contexts. For example:
These are just some of the difficult operational questions, but they do not compromise the utility of the theoretical framework. Although incomplete, this framework is a useful conceptual tool to address policy questions about why interventions work in some areas and not in others. It provides clues as to how to modify interventions and increase the likelihood of success.
Anderson, W.L. 1914. The country town: a study of rural evolution. New York, Doubleday, Page and Company.
Arefeen, H.K.S. 1983. Changing agrarian structure in Bangladesh: Shimulia: a study of a peri-urban village. Dissertation Abstracts International, A 1984, 44(10): 3104. Syracuse University, New York, United States. 1983, 222 pp.
Bettison, D.G. 1958. The social and economic structure of a sample of peri-urban villages, Blantyre-Limbe, Nyasaland. Rhodes Livingston Communications No. 12. Lusaka, Zambia, Rhodes Livingston Institute. 95 pp.
Cadene, P. 1990. Using the space around urban areas: a regional geography of conflict. Étude rurales: revue trimestrielle d'histoire, géographie, sociologie et économie des campagnes, 118-119: 235-267. Paris, Mouton and Co.
Clauss, S. 1991. The economic importance of women in West African rural areas and their role in the development process. In Diarra, ed. Westafrika zwischen autochthoner Kultur and Modernisierung. Ein Afrika-Reader, p. 241-262. Frankfurt, Germany, Aachener Beitrage zur vergleichenden Soziologie und zur China Forschung No. 7. (in German)
Clough P. 1996. Planning and sustainable management: a re-examination of the peri-urban problem. Review of Marketing and Agricultural Economics, 64(3): 277-289.
Crankshaw, O. 1993. Squatting, apartheid and urbanisation of the Southern Witwatersrand. African Affairs, 92(366): 31-51.
Cribier, F. 1973. The second homes of urbanites in the French countryside. Étude rurales: revue trimestrielle d'histoire, géographie, sociologie et économie des campagnes, 49-50: 181-204. Paris, Mouton and Co.
Crowley, E. & Appendini,
K. 1999. Rural poverty: population dynamics, local institutions and access
to resources. Paper presented at the Thematic Workshop on Population, Poverty
and Environment held at FAO, Rome, October 1998.
Available from: www.fao.org/waicent/faoinfo/sustdev/wpdirect/wpre0091.htm
de Soto, H. 1989. The other path: the invisible revolution in the Third World, translated by J. Abbott. New York, Harper and Row. 271 pp.
FAO. 1999a. Urban and peri-urban food production: a new challenge for the Food and Agriculture Organization (FAO) of the United Nations, by A.W. Drescher and D.L. Iaquinta. Internal report. Rome. 67 pp.
FAO. 1999b. Urban and peri-urban agriculture. 15th session of the Committee on Agriculture (COAG), Rome.
FAO. 1999c. The impact of ageing populations on food security: the role of intergenerational transfers and land tenure dynamics, by D.L. Iaquinta. Internal report. Rome.
Fischer, C. 1977. Networks and places: social relations in the urban setting, New York, Free Press. 229 pp.
Fischer, C. 1984. The urban experience. Second edition. San Diego, California, United States, Harcourt Brace. 371 pp.
Friedmann, J. 1996. Modular cities: beyond the rural-urban divide. Environment and Urbanization, 9(1): 129-131.
Freidman, J. & Wolff, G. 1982. World city formation: an agenda for research and action. Los Angeles, California, United States, University of California Press.
Gans, H.J. 1962. The urban villagers : group and class in the life of Italian-Americans. New York, Free Press. 367 pp.
Gardner, K. 1993. Desh-bidesh: Sylheti images of home and away. Man, 28(1): 1-16.
Groppo, P. & Toselli, P. 1997. Desde la ciudad al territorio: La nueva problemática peri-urbana. In FAO. Land reform: land settlement and cooperatives, 1997/2: 5-20. Rome.
Hewitt, M. 1989. Defining "rural" areas: impact on health care policy and research. Washington, DC, Health Program, Office of Technology Assessment, Congress of the United States.
Holland, M.D., Kasanga, R.K., Lewcock, C.P. & Warburton, H.J. 1996. Peri-urban baseline studies, Kumasi, Ashanti Region, Ghana. Volume 1: Executive summary and main report. United Kingdom, Natural Resources Institute, Overseas Development Administration.
Holleman, J.F. 1964. Experiment in Swaziland: report of the Swaziland Sample Survey 1960 by the Institute for Social Research, University of Natal for the Swaziland Administration. London, Oxford University Press. 348 pp.
Julliard. 1973. Urbanization of the countryside. Étude Rurales: revue trimestrielle d'histoire, géographie, sociologie et économie des campagnes, 49-50: 5-9. Paris, Mouton and Co.
Mandel, R. 1990. Shifting centres and emergent identities: Turkey and Germany in the lives of Turkish Gastarbeiter. In D. Eickleman and J. Piscatori, eds. Muslim travellers, London, Routledge.
Mathews, J.S. 1941.The peri-urban fringe of Akron, near Copley, Ohio. Kent State University, Ohio, United States. 112 pp. (M.A. thesis)
Mayer, P. 1961. Townsmen or tribesmen. Oxford, United Kingdom, Oxford University Press.
McDowell, C. & de Haan, A. 1997. Migration and sustainable livelihoods: a critical review of the literature. Institute of Development Studies Working Paper No. WP65. Brighton, United Kingdom, University of Sussex. 28 pp.
OECD. 1979. Agriculture in the planning and management of peri-urban areas. Volume 1: synthesis. Paris. 94 pp.
Rambaud, P. 1973. Village and urbanization: sociological problems. Étude Rurales: revue trimestrielle d'histoire, géographie, sociologie économie des campagnes, 49-50: 14-32. Paris, Mouton and Co.
RUPRI. 1998. Defining "rural": definitions of rural areas in the US. Rural Policy Context, Rural Policy Research Institute at: www.rupri.org/policyres/context/rural.html
Schlyter, A. 1991. Twenty years of development in George, Zambia. Stockholm, Swedish Council of Building Research.
Sonje, V. & Stulhofer, A. 1995. Not so dangerous liaisons between sociology and economics: toward a socio-economic model of institutional change. Revija Za Sociologiju, 26(3-4): 189-202.
Stahl, H.H. 1973. Urbanization and rural life in Romania. Étude Rurales: revue trimestrielle d'histoire, géographie, sociologie et économie des campagnes, 49-50: 296-303. Paris, Mouton and Co.
Starosta, P. 1993. Institutional conception of changes in rural local communities. Przeglad Socjologiczny, 40(1): 57-71.
United States Census Bureau. 1999. Appendix A: Area Classifications. Available from: www.census.gov/td/stf3/append_a.html#urbanandrural
Vele, M. 1978. Rural village and peri-urban settlement: a case study of circulation from the Central Province. Occasional Paper No. 2. Institute of Applied Social and Economic Research. 25 pp.
von Thünen, J.H. 1848. Der isolierte Staat in Beziehung auf Landwirtschaft und Nationalökonomie. Rostock, Germany, G.W. Leopold's Universitatsbuchhandlung, 1842-1863.
von Thünen, J.H. 1966. Isolated state, English edition of von Thünen, 1848, translated by C.M. Wartenberg. Oxford, United Kingdom and New York, Pergamon Press. 304 pp.
Whisson, M.G. 1984. Tsweletswele: from settlement to community in a peri-urban area of the Ciskei. Urban Anthropology, 13(2-3): 237-259.
Williams, J.F., Brunn, S.D. & Darden, J.T. 1983. World urban development. In S.D. Brunn and J.F. Williams, eds. Cities of the world: world regional urban development, p. 3-42. New York, Harper Collin Publishers.
1 Portions of this paper were developed with support from FAO's Partnership Programme. The results are part of a spontaneously generated collaborative project carried out by the authors on the subject of urban and peri-urban agriculture (UPA). David L. Iaquinta was sponsored by the Land Tenure Service (SDAA) in cooperation with Nebraska Wesleyan University (United States). Axel W. Drescher was sponsored by The Rural Institutions and Participation Service (SDAR) in cooperation with Freiburg University (Germany).
2 Throughout the paper, the term UPA is used to refer collectively to agricultural production - including horticulture, floriculture, animal husbandry, forestry and fisheries - in both urban and peri-urban environments.
3In their review of the literature, the authors were unable to identify the original source of the term peri-urban. However, most earlier references use it in the context of an applied category rather than a theoretical concept. See, for example, Bettison, 1958; Holleman, 1964; and Mathews, 1941.
4See, for example, FAO, 1999b. Whether acknowledged or not, all such research derives from the ideas of von Thünen (1848, and 1966), which are often put into operation by demographic indicators (e.g. population size and/or density).
5For example, Hewitt (1989) argues that rural itself is not a "single category but a complex continuum. Rural areas share the common characteristics of comparatively few people living in the area, limited access to large cities (and sometimes even to smaller towns), and considerable travelling distances to `market areas' for either work or everyday living activities. They exist along a continuum, however, from more rural to less rural and vary extensively based on the following factors: 1) proximity to a central place, 2) community size, 3) population density, 4) total population, and 5) economic/socio-economic factors."
6Julliard (1973) argues that urbanization of the countryside (integration of rural inhabitants into new economic and social relations with town-dwellers) can be interpreted either as the obliteration of the countryside or as the cooperation of rural and urban inhabitants, resulting in the disappearance of the town-country dichotomy.
7Stahl (1973) examines Romania and concludes that disparities between towns and villages are disappearing, not because villages are becoming towns but because they no longer suffer from social and economic under-develop-ment. These changes are arising from a combination of urbanization, deruralization (i.e. urbanism as defined in this article) and modernization (of agricultural techniques and the formation of farming cooperatives) which alter both the composition of the rural population and the structure of the agricultural family.
8It could equally well start with the concept of rural, although this is problematic because, on the whole, the urban has exercised an intellectual hegemony in the minds of researchers.
9This is not to imply that the "mental attitude" of the peri-urban dweller is identical to that of the urban dweller or that it automatically excludes significant elements of the "rural mental attitude". Drescher's own experience from Lusaka suggests that peri-urban is quite rural but its production is directed to the urban environment. It is this very range of attitudes that causes the variation of peri-urban types introduced in the next section of this article.
10Holleman (1964: 333) emphasizes the importance of this social-psychological component when referring to "the very nature of peri-urban settlement is that, to a rural-derived but urban-oriented people, it appears to offer the best of two worlds." (Emphasis added.)
11For example, Groppo and Tosselli (1997) identify the system of "external" urban values as a decisive factor in defining urban and peri-urban.
12As McDowell and de Haan (1997: 9) point out: "Migration studies is not just about movement, but also the interconnectedness of place of origin and place of destination." Mandel (1990), quoted in Gardner (1993: 11), adds that "migration is essentially a series of exchanges between places."
13Rambaud (1973) hints at this process when he defines urbanism as the creation and modelling of a space where a group can express itself. He points out that this process always took place at the village level but that the form it previously took is being changed or copied and deprived of its functions by urbanization. Thus, village urbanization is to be seen as only part of town development. Presumably, out-migrants are then the agents of this broader form of urbanization which is introduced to the village.
14For example, Crankshaw (1993) studied the rapid growth in South African squatter settlements and found that they had not resulted from the uncontrolled African urbanization that followed the abolition of influx control (pass) laws. Instead the primary impetus came from the displacement of workers from peri-urban farms. When small settlements of displaced farm workers started to grow, news spread and displaced urban-dwellers began to seek refuge in these settlements as well.
15Whisson (1984) describes one example of this process: Tsweletswele was an informal settlement in the Ciskei "homeland" of the Eastern Cape Province of South Africa. It comprised a large number of rootless people, most of whom were unskilled and illiterate, who had come from many places for many reasons, had no legal right to stay and whose presence was inconsistent with the regional planing of the Ciskei authorities. Yet this amorphous settlement was transformed into an ordered community by the development of various social and cultural elements. Economic activities, kinship links and rituals generated relationships and a communal spirit. However, care should be taken in applying these results elsewhere because of the unique conditions in South Africa both under apartheid and after its collapse.
Thus, 16not only do the institutions and networks vary by peri-urban type but their relative importance to community development varies systematically as well. This is an important point for development workers. It argues for the clear a priori identification of which institutions need to be promoted in each type of environment.
17Observations of Schlyter (1991) show that this is also reflected in the settlement pattern of illegal squatter compounds. This pattern seems to be derived from rural settlement, but the scale is enlarged and the social content is different. Schlyter's interviews clearly indicate that people did not try to reconstruct their communities of origin but were aiming to recreate what they saw as an urban life style.
18These areas seem closest to those advocated by Friedmann (1996) as appropriate focuses for development. His theoretical model of modular urbanization envisions town-centred, self-governing agropolitan districts and calls for the development of high-density rural or peri-urban areas to raise living standards and increase employment opportunities. He suggests that agropolitan districts would preserve the integrity of households and village communities, thus reducing the scale of migration to cities and the social costs inherent to urban-based development.
19Cadene (1990) examines this conflict process in the rural peripheries of ten large French cities. Here urban newcomers construct private houses on formerly agricultural lots, while the agricultural activity of the area generally remains dynamic. Cadene identifies three types of conflicts: territory management, usage of communal space, and urbanization of communal space. However, not all such interaction results in conflict. Earlier work in France by Cribier (1973) showed that relations between the owners of second (i.e. country) homes and local residents depend on the socio-economic situation of the former and the traditions of hospitality of the latter.
20There are some links between the notions underlying the concept of APU used in this article and Mayer's description of the red Xhosa in East London, South Africa as "the `encapsulated' traditionalist whose entire urban tenure, regardless of length, is devoted to a single aim: to gather the financial means for an ultimate, undisturbed retirement at his rural home in the society whose morals and values he has never forsaken." (Mayer as cited in Holleman, 1964: 332.)
21 Holleman (1964) refers to this as the "dualistic basis of existence, both `rural' and `urban'." His evidence from Swaziland shows that "in the urban sector alone nearly half of the adult male population considers it has substantial roots in the rural area.... The same tendency is clearly revealed among the peri-urban males, but in this sector the proportions of those with a dual home basis are very much smaller than in the urban sector."
22Vele (1978) describes an example from the Central Province of Papua New Guinea. Focusing on the role of circulation among the rural village, a peri-urban settlement (either chain or diffuse PU) and the city of Port Moresby, he argues that contemporary circulation is not simply an extension of historical movements but has emerged in its present form as a result of the introduced influences of education, waged employment, better communication and urbanization.
23Consider the complicated relationships in West African cultures that link geographic environments through gendered family relationships. Here the migration of large numbers of young men to the cities has often left women wholly responsible for all agricultural and family maintenance work. Yet, despite the fact that African women provide 80 percent of the (mostly unpaid) labour on the land and have traditionally grown the food crops, they have rarely been given adequate access to innovations, development and credit (Clauss 1991).
24In post-communist Poland, Starosta (1993) asserts that rural transformations are caused by the replacement of traditional informal institutions with new supralocal formalized institutions, conducive to the formation of new groups of structures with overlapping spatial coverage. He argues that the concentration of formal institutions at the local district (commune) level represents the emergence of a new type of rural territorial aggregation. See FAO (1999c) for a discussion relating migration to population ageing and intergenerational transfers.
25 One such sequence is the transition of a "place" from rural to village peri-urban to absorbed peri-urban. An example of this process is the village of Shimulia near Dhakar in Bangladesh, described by Arefeen (1983). While not a perfect example, this case manifests some of the essential features of the transition.
26See, for example, Crow*ley and Appendini (1999) for a description of the cargo system in Oaxaca, Mexico.
27The term "defensive insulation" derives from subgroup theory; see Fischer (1984).
28Additional support for the differential resource classification of chain and in-place PU, relative to absorbed PU, on the one hand, and village PU, on the other, comes from work by Holleman (1964): "Comparatively speaking, it appears that peri-urban dwellers acquire their homes sooner than urban dwellers, and on the whole the former group has a slightly higher proportion of house-owners. The evidence with regard to home-ownership is, however, far from conclusive.... [A] substantial proportion of wage-earners have found it necessary to provide some sort of accommodation for themselves and their dependants, without necessarily committing themselves to a permanent urbanized existence.... [T]he physical development of these urbanized settlements has on the whole been so haphazard that there is in the majority of cases no certainty about the rights pertaining to the individual plots upon which these structures have been built." (p. 330) "Therefore, whatever value may be attached to urban home-ownership as a factor in the process of urbanization in Swaziland at the present time, it does not appear to interfere with the retention of strong ties with the rural home area."
29It is also possible for a phenomenon to be legitimate if it is endorsed by those in a population who are charged with making and enforcing rules, even if the population at large does not endorse it.